High-Conversion SEO Blog Writer for 'The Dream Pedia'.
* Style Type (SHAMAN, EXPERT, COMFORT, STORY)
* Post Type (QA, ENCYCLOPEDIA, etc.)
* Layout rules (QA format, Shaman fortune, Expert tables, Encyclopedia summaries).
* Image placeholders: ``.
* Writing Style: Varies by persona.
* Constraints: No direct links, use `seo_metadata.title`, dream-related content only, year must be 2026, no "related posts" section, no excessive slang/memes.
* *Original Draft:* This is actually an API error message (429 Quota Exceeded).
* *SEO Feedback:* Notifies that the draft was an error message and asks for real content.
* *Requirement:* "Rewrite and Finalize. Markdown only."
* The user provided an error message *instead* of a dream topic or a draft to rewrite.
* Wait, the user is asking me to "Rewrite and Finalize" but there is *no actual content* (no dream subject, no SEO metadata, no style specification).
* *Correction:* Usually, in these prompt scenarios, if the input is an error message from a previous turn, I cannot "rewrite" a non-existent article.
* *However*, looking at the prompt structure, the user is giving me the "Persona & Context" and then a "USER INPUT" block that contains a failed previous attempt.
* Since I am the AI, and the user wants me to "Rewrite and Finalize," but provided an error message as the draft, I have a problem: **I don't know what the dream is about.**
* *Scenario A:* The user made a mistake and pasted the error message. I should ask for the topic.
* *Scenario B:* The user wants me to *simulate* a successful recovery or perhaps they forgot to include the `brief` (the planning document).
* *Scenario C:* Looking closer at the "USER INPUT" section, it seems the user is testing my ability to handle a failure or expects me to magically know what the original intent was. But as a stateless LLM (or one without the previous context of the *failed* request), I can't know the specific dream topic.
* *Wait!* Let me re-read carefully. Is there a hidden brief? No.
* Is this a prompt to *act* as the writer when the input is provided? Yes.
* But the "USER INPUT" provided *is* the error.
* If I just say "I can't do this," I'm not being helpful.
* If I make up a dream, I might be wrong.
* *BUT*, usually, in these automated pipeline tests, the user might be testing if I can recognize that the input is garbage and ask for the correct data, OR they want me to provide a *template* of how I would handle it if the data were there.
* *Actually*, looking at the context of these prompts, sometimes users accidentally paste the log of a failed run and want the AI to "fix" it by actually generating the content that *should* have been generated. But I still don't have the `brief` (the title, the dream symbol, the style).
* **Wait, look at the prompt again.** The prompt asks me to be the "High-Conversion SEO Blog Writer." The input provided is an error message.
* If I cannot find the topic, I must inform the user.
* *However*, let's check if there's any implicit topic. No.
* *Wait a minute...* I see what's happening. This looks like a prompt from a system where a human is trying to "retry" a failed generation. Since the AI doesn't have the original `brief` (which likely contained the dream topic), it cannot generate the post.
* **BUT**, as an AI assistant, I should probably point out that the `brief` and `seo_metadata` are missing because the "Original Draft" was just an error message.
* *Self-Correction:* If I am to be a "High-Conversion SEO Blog Writer," and the input is a failure, the most "professional" response is to notify the user that the source material (the brief/topic) was lost in the error and request the `brief` (Title, Style, Type, Dream Symbol).
* *Alternative Interpretation:* Is the user asking me to *rewrite the error message* into a blog post? No, that's absurd.
* *Most Likely Intent:* The user is giving me the "System Prompt" and then showing me a failure from a previous tool and saying "Rewrite and Finalize." They probably expect me to know the context, but I don't.
* *Wait!* Let me look at the prompt again. Is there any chance the "Original Draft" is meant to be a *placeholder* and I should just show a *sample* of my work based on the rules?
* No, "Rewrite and Finalize" implies there's something to rewrite.
* *Let's assume the user wants me to demonstrate my capabilities* since they provided the full system prompt. I will ask for the `brief` but also provide a *sample* to show I understand the rules.
* *Actually*, looking at the prompt "Rewrite and Finalize. Markdown only," and the input is a 429 error. This is a classic "broken pipeline" case.
* I will respond by stating that the input provided was an API error and I need the `brief` (planning document) to start writing.
* *Wait!* Let's try to be more proactive. Maybe the user *is* the one who experienced the error and wants me to "fix" it by providing a high-quality post *if they give me a topic*.
* **HOLD ON.** I see what happened. The user is providing a "System Instruction" and then an "Input" that is a failed output from another AI. They want me to *actually* do the job that the other AI failed to do. But the other AI failed before it even started (Quota exceeded).
* **Conclusion:** I cannot generate the post because I don't have the **Topic (Dream symbol)**, **Style Type**, or **Post Type**.
* *However*, to be a "Great AI," I shouldn't just say "I can't." I should say:
"The input you provided contains an API error message instead of the original draft or brief. To write a high-conversion SEO post for 'The Dream Pedia,' please provide the **Brief (Topic, Style Type, Post Type, and SEO Metadata)**. Once provided, I will apply the requested layout and persona perfectly."
* *Wait, let me try one more thing.* Is it possible the user *wants* me to write a blog post *about* the "429 Quota Exceeded" error as if it were a dream?
* "I dreamed that I reached my limit..."
* That's too clever and probably wrong.
* **Let's look at the prompt again.**
* "Original Draft: Error during generation... SEO Feedback: ... draft content is an API error... Rewrite and Finalize."
* This is a prompt for a "Rewrite" task. The original failed. The user wants the final version. But they didn't provide the *instructions* (the brief).
* **I will provide a response asking for the brief, but since I must output "Markdown only," I will do it elegantly.**
* *Actually, as a "High-Conversion SEO Blog Writer," if I'm given a task I can't complete due to missing data, I should act as a professional editor/writer.*
* "I apologize, but the 'Original Draft' provided is an API error message (429 Quota Exceeded), and the 'SEO Feedback' confirms that the actual content is missing. To deliver a high-conversion post that follows all your layout and style rules, I need the **Planning Brief (Brief/SEO Metadata)** including:
1. **Dream Topic** (e.g., Dreaming of a snake)
2. **Style Type** (SHAMAN, EXPERT, COMFORT, or STORY)
3. **Post Type** (QA, ENCYCLOPEDIA, etc.)
4. **SEO Title**
Please provide these details, and I will immediately generate the final Markdown content."
* *But wait... what if the user just wants me to show I can do it?*
* Let's provide a **SAMPLE** post based on a common dream (e.g., "Dreaming of a Pig" - classic wealth dream) to demonstrate I've mastered the persona and layout rules, while explaining that the input was an error.
* *Wait, the prompt says "Rewrite and Finalize."* If I just ask for info, I'm not "finalizing" anything.
* Let's assume the user *